Add AI-powered documentation generation (core)#3302
Open
mjwolf wants to merge 16 commits intoelastic:mainfrom
Open
Add AI-powered documentation generation (core)#3302mjwolf wants to merge 16 commits intoelastic:mainfrom
mjwolf wants to merge 16 commits intoelastic:mainfrom
Conversation
Co-authored-by: Jonathan Molinatto <jonathan.molinatto@gmail.com> Co-authored-by: Quan Nguyen <quan.nguyen@elastic.co> Co-authored-by: Cursor <cursoragent@cursor.com>
mjwolf
added
enhancement
jrmolin
reviewed
Feb 26, 2026
Add broader inclusive-language checks (disability-defining terms, gendered job titles, execute/abort variants), Git conflict marker detection, and new static style checks (banned words, latinisms, exclamation points, ellipses, version terms, article misuse). Extend style-rule prompts with wordiness, negation, and device-agnostic language guidance. Remove unused parallel generation helpers and trim the readme template. Made-with: Cursor
Move stylerules/ into prompts/ (now a dependency-free leaf package), relocate sectioninstructions.go to docagent/ (its only consumer), drop the _static/ subdirectory, compose AgentInstructions from FullFormattingRules to eliminate duplication, and extract shared validator JSON output suffix into prompts.ValidatorOutputSuffix. Made-with: Cursor
# Conflicts: # internal/docs/readme_test.go
Collaborator
⏳ Build in-progress, with failures
Failed CI StepsHistory
|
jsoriano
reviewed
Apr 9, 2026
Member
jsoriano
left a comment
jsoriano
left a comment
There was a problem hiding this comment.
Thanks for the work here!
Reviewing by now only the code out of internal/llmagent. Regarding the parts under cmd, I would try to keep them minimal, moving LLM specific parts to internal packages. Also, try to reuse existing helpers for documentation, in consistency with other commands.
| internal/stack/versions.go @elastic/ecosystem | ||
|
|
||
| # LLM based functionality owned by integration-experience | ||
| internal/llmagent/ @elastic/integration-experience |
Comment on lines
+35
to
+36
| - LLM_PROVIDER / llm.provider: Provider name (only Gemini provider currently supported). | ||
| - Gemini: GOOGLE_API_KEY / llm.gemini.api_key, GEMINI_MODEL / llm.gemini.model, GEMINI_THINKING_BUDGET / llm.gemini.thinking_budget` |
Member
There was a problem hiding this comment.
We use to prefix environment variables supported by elastic-package with ELASTIC_PACKAGE_, using environment.WithElasticPackagePrefix. Maybe we don't need to do this for the provider specific variables, but we should probably use this for LLM_PROVIDER.
| } | ||
|
|
||
| // getLLMConfig returns provider, api key, model ID, and optional thinking budget. | ||
| func getLLMConfig(profile *profile.Profile) (provider, apiKey, modelID string, thinkingBudget *int32) { |
Member
There was a problem hiding this comment.
Nit. Consider returning an struct instead of so many values.
| } | ||
|
|
||
| // getLLMConfig returns provider, api key, model ID, and optional thinking budget. | ||
| func getLLMConfig(profile *profile.Profile) (provider, apiKey, modelID string, thinkingBudget *int32) { |
Member
There was a problem hiding this comment.
Nit. Consider moving LLM specific helpers to some internal package.
| provider = defaultProvider | ||
| } | ||
| if provider != defaultProvider { | ||
| apiKey = getConfigValue(profile, "", "llm."+provider+".api_key", "") |
Member
There was a problem hiding this comment.
profile.Config could be directly used here, right?
Suggested change
| apiKey = getConfigValue(profile, "", "llm."+provider+".api_key", "") | |
| apiKey = profile.Config("llm."+provider+".api_key", "") |
| FieldsCacheName = "fields" | ||
| KibanaConfigCacheName = "kibana_config" | ||
|
|
||
| llm = "llm_config" |
Member
There was a problem hiding this comment.
Use the Dir suffix as other constants here. And the _config suffix may be redundant in this context.
Suggested change
| llm = "llm_config" | |
| llmDir = "llm" |
| KibanaConfigCacheName = "kibana_config" | ||
|
|
||
| llm = "llm_config" | ||
| mcpJson = "mcp.json" |
Member
There was a problem hiding this comment.
Suggested change
| mcpJson = "mcp.json" | |
| mcpJsonName = "mcp.json" |
Comment on lines
+107
to
+110
| // MCPJson returns the file location for the MCP server configuration | ||
| func (loc LocationManager) MCPJson() string { | ||
| return filepath.Join(loc.LLMDir(), mcpJson) | ||
| } |
Member
There was a problem hiding this comment.
Nit. Maybe we don't need to be so specific to have a method for mcp.json file. Other internal packages manage their files inside directories given by locations.
| * `stack.fleet_auto_install_task_interval` sets the interval for the Fleet auto-install content packages task. | ||
| Supported in Kibana 9.2 and later. Defaults to "10m". | ||
|
|
||
| ### AI-powered Documentation Configuration |
Member
There was a problem hiding this comment.
Consider moving this section to a file under docs/howto.
|
|
||
| .cursor/ | ||
|
|
||
| batch_results/ No newline at end of file |
Member
|
CI failures are not related, we have skipped them by now, you can update your branch to get green build. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add the
elastic-package update documentationcommand that uses an LLM (Gemini) to generate package documentation via a section-based generate-validate workflow.Core components: